It is obvious that digital solutions will be useful for crisis management due to the amount of data to manage and share in a hurry. Thus, lots of R&D projects proposed solutions to help the stakeholders. Nevertheless, the transfer from the results of R&D projects to stakeholders is difficult due to mainly the maturity of the solutions and the lack of a unified perception of the situation. This lack is due to the fact that a constellation of solutions is needed in crisis management. Based on these observations, we start developing during the Criz'Innov French project, a middleware, named CrizLABTM, which aims to facilitate the transfer from the results of R&D projects in crisis management to stakeholders or industries. Therefore, the middleware allows to:
- Tests the results in various crisis scenarios
- Facilitate the interactions between results and existing solutions in crisis cells
- Support for demonstration
To achieve these objectives, CrizLABTM needs to be able to support a unified approach of interoperability based on a common vocabulary, taxonomy. Thus, each result keeps its own way to structure the data and a gateway is used to transform the exchanged data from the specific model to the unified and shared way to describe the data. Based on previous experiences, it seems that the unified approach of interoperability is relevant only regarding the exchange of data between applications. Furthermore, crisis management is mainly based on the occurrence of events. Therefore, CrizLABTM is based on EDA (Event-Driven Architecture) and proposes the following functionality:
- Manage the COP (Common Operational Picture): COP is an overview of the situation, generally showing on a map. This COP is modified on each received event.
- Manage the history of the situation: the history of the situation is saved and accessible for humans as well as other applications.
- Manage the link between applications. Several applications have to be executed in sequence in order to make advice or provide an added value information
Our team also explore new modes of representation and interaction, particularly on web technologies (Dashboard, VR / AR, Tangible objects, large tactile table…), enabling dynamic and real-time visualization of massive datasets. The connected visualization interfaces linked to CrizlabTM need to be improved, particularly user interfaces showing the outcomes of an AI across multiple probable future scenarios and the related response strategies.